The Lyceum: Space Economy Weekly — Aug 13, 2026

Space hardware delivered a clarifying week. China lost a rocket, Rocket Lab’s satellite business outran its next launch vehicle, and SpaceX showed that constellation scale increasingly depends on parallel operations—not a single spectacular flight. Meanwhile, Europe and U.S. regulators advanced two less photogenic but essential pieces of space infrastructure: sovereign connectivity and spectrum ownership.

• Starlink 17-49 (SpaceX): SpaceX deployed 24 Starlink satellites from Vandenberg Space Force Base on August 11, adding capacity to its low-Earth-orbit broadband network.
• EO2GO: Wildfire Edition (EO2GO): Copernicus Data Space added a mobile wildfire service that packages satellite-derived fire locations, perimeters, vegetation conditions, and weather information for field use, including in low-connectivity environments.

China’s Long March 7A failed shortly after lifting off from the Wenchang Space Launch Site on August 10, damaging the rocket and its communications-satellite payload, according to the Associated Press. China Aerospace Science and Technology Corporation has opened an investigation.

The Long March 7A carries heavier spacecraft toward high-energy destinations such as geostationary transfer orbit—the route satellites use to reach positions roughly 36,000 kilometers above Earth. This was more than a lost mission. Every payload using the same vehicle, engine family, or suspect component may now face a schedule review.

If investigators isolate the problem quickly, China can treat the loss as a contained manufacturing or quality-control failure. If related launches are paused, the effects spread into government communications programs and an already crowded global market for geostationary launch capacity. The signal to watch is not the investigation’s opening, but which other missions leave the calendar while it runs.

Rocket Lab’s satellite business is moving faster than its launch ambitions. The company reported record second-quarter revenue of $234 million and a $2.36 billion backlog. Its space-systems work now includes a contract worth up to $397 million for multiple U.S. Space Force satellites and a Neutron launch, plus more than $160 million to build three geostationary spacecraft.

Neutron, however, is moving more slowly. Rocket Lab said its replacement first-stage tank remains aligned with delivery to the company’s Virginia launch site in the fourth quarter. Delivering a rocket to the pad is not the same as launching it, leaving little schedule margin for a 2026 debut.

If Neutron works, Rocket Lab can offer customers the spacecraft, launch, and—in time—communications infrastructure inside one corporate stack. If it slips deeply into 2027, Rocket Lab can still grow as a satellite prime contractor, but it will keep buying launch access from the market it wants to disrupt. Tank qualification and physical arrival in Virginia will separate an executable schedule from a hopeful one.

Europe’s sovereign-connectivity project is now a 348-satellite delivery challenge. The European Union’s €15.6 billion IRIS² secure-connectivity program has moved into implementation with a revised architecture of 348 satellites, according to Tom’s Hardware. Initial services are targeted for 2029; no IRIS² spacecraft are operating yet.

IRIS² is not primarily a consumer-broadband attempt to out-Starlink Starlink. Its first job is encrypted connectivity for European institutions, military users, emergency services, and critical infrastructure. The expanded design combines low Earth orbit, which offers shorter signal paths, with medium Earth orbit, where each satellite can cover a wider area.

Success would give Europe a communications layer it can control during crises—and create a large block of demand for European spacecraft, terminals, gateways, and launch providers. Failure looks less dramatic: procurement fragmentation, incompatible national requirements, and a 2029 service date that quietly becomes aspirational. Firm manufacturing and launch awards will show whether the architecture has become an industrial schedule.

Firefly has sold Lockheed Martin access to a launch queue. Now it must build the queue. Firefly Aerospace extended its Lockheed Martin agreement to cover as many as 25 Alpha launches through 2031. These are options rather than 25 fully scheduled flights, and Firefly has not disclosed their payloads, dates, or destinations.

The missions are intended to use Alpha Block II, a taller configuration with redesigned avionics, batteries, thermal protection, and composite structures. Firefly also reported $117.7 million in quarterly revenue and a $92.3 million net loss as it continued spending on development and manufacturing.

If Block II flies reliably, Lockheed Martin gains control over deployment timing and Firefly gains something small launch badly needs: repeat demand that can support a production line. If Block II’s debut slips or early flights stumble, those options remain paper capacity rather than revenue-generating backlog. A named payload and firm first-flight date are the next evidence that Firefly has sold a cadence, not merely the possibility of one.

SpaceX launched 53 Starlink satellites on August 11—and the total was only part of the story. The company deployed 29 satellites from Florida and 24 from California, according to reports collected in the research. The critical detail was the use of separate coasts, ranges, pads, recovery operations, and processing teams on the same day.

That operational parallelism turns launch cadence into network capacity. It also lets SpaceX replenish older spacecraft, fill coverage gaps, and expand service without depending on a single range or launch complex.

If this tempo becomes routine, Starlink’s advantage will come as much from logistics as satellite design—and outside customers may find themselves competing with SpaceX’s own constellation for favorable manifest positions. Failure would appear as pad congestion, booster shortages, or repeated schedule movement for third-party payloads. Watch Vandenberg in particular: polar-orbit customers have fewer practical alternatives when its calendar tightens.

A new preprint puts a number on the interference risk from large constellations. Submitted on August 12, it proposes per-satellite radiation and spurious-emission limits across 1–14 gigahertz. Using measurements from the SNIFFLES-I survey, the authors modeled interference from roughly 12,000 catalogued spacecraft and considered the much larger populations described in constellation filings.

For one direct-to-device downlink at 2.62 gigahertz, the authors calculate that a harmonic—an unwanted signal at a multiple of the intended frequency—could cause at least 59% data loss for geodetic very-long-baseline interferometry on the session. That technique uses widely separated radio telescopes to maintain precise terrestrial and celestial reference frames.

The paper is a preprint, not a peer-reviewed standard or regulatory requirement. But if regulators adopt its basic approach, constellation licensing could move from broad operator promises to emissions tests for every spacecraft leaving the factory. Non-adoption would leave observatories negotiating constellation by constellation. The decisive signal will be whether national regulators or the International Telecommunication Union begin requesting per-satellite verification data.

Rocket Lab’s proposed $8 billion acquisition of Iridium has entered the regulatory plumbing. Rocket Lab said on August 13 that the U.S. antitrust waiting period had expired. Rocket Lab and Iridium have also filed transfer applications with the Federal Communications Commission; the transaction remains subject to additional approvals and closing conditions.

The strategic asset is not only Iridium’s satellites. It is the company’s globally used L-band network, whose lower-frequency signals are valued for resilient mobile and safety-related communications. Owning that network would move Rocket Lab from building and launching spacecraft into operating communications infrastructure.

An uncomplicated approval would validate Rocket Lab’s effort to assemble launch, satellite manufacturing, and recurring services under one roof. Conditions involving spectrum, cybersecurity, continuity of service, or government access would reveal that Washington sees Iridium as strategic infrastructure rather than an ordinary telecom asset. The next meaningful evidence will be the FCC’s questions and conditions—not another acquisition presentation.

A Chinese communications satellite disappeared into a fireball, Rocket Lab carried an $8 billion spectrum dowry into Washington, and two Falcon 9 teams spent August 11 tossing routers over opposite oceans.

The Moon may be high ground, but the Madrid firebreak and the FCC filing queue are where space infrastructure discovers gravity.

Keep watching the plumbing.

Forward this to someone who still thinks the space economy happens entirely above the atmosphere.